ClawHub
Back to skill

Security audit

Agent Creation

Security checks for vulnerabilities and agentic risk

Overview

This instruction-only skill creates local OpenClaw agent workspaces as advertised, with no evidence of hidden, unrelated, or deceptive behavior.

Install this only if you want help creating persistent local OpenClaw agents. Before running its suggested commands, confirm the agent name and workspace path, check whether the target workspace already exists, and review the generated SOUL.md and memory file content.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The trigger conditions are broad and include common phrases like "new agent" and "configure agent," which can cause the skill to activate in situations where the user did not explicitly intend agent creation. Because this skill performs workspace creation and configuration actions, unintended activation can lead to unauthorized local state changes and accidental agent provisioning.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill instructs the agent to replace SOUL.md and create memory files without warning the user that local files will be created or overwritten. This is dangerous because users may not realize the skill performs persistent filesystem modifications, increasing the risk of accidental data loss or unexpected changes to agent configuration.

Static analysis

No suspicious patterns detected.